Minimum distance computation of LDPC codes using a branch and cut algorithm

Ahmet B. Keha, Tolga M. Duman

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

We give a branch-and-cut algorithm for finding the minimum distance of a binary linear block code. We give two integer programming (IP) models and study the convex hull of the single constraint relaxation of these IP models. We use the new inequalities as cuts in a branch-and-cut scheme. Finally, we report computational results based on turbo and low density parity check (LDPC) codes that demonstrate the effectiveness of our cuts. We demonstrate that our IP formulation and specific cuts are efficient tools for determining the minimum distance of moderate size linear block codes, specifically, they are very efficient for LDPC codes, and provide us with an additional tool for solving this important problem.

Original languageEnglish (US)
Article number5439310
Pages (from-to)1072-1079
Number of pages8
JournalIEEE Transactions on Communications
Volume58
Issue number4
DOIs
StatePublished - Apr 2010

Fingerprint

Integer programming
Block codes

Keywords

  • Branchand-cut
  • Linear block codes
  • Low density parity checkcodes
  • Mixed-integer programming modelling
  • Turbo codes

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Minimum distance computation of LDPC codes using a branch and cut algorithm. / Keha, Ahmet B.; Duman, Tolga M.

In: IEEE Transactions on Communications, Vol. 58, No. 4, 5439310, 04.2010, p. 1072-1079.

Research output: Contribution to journalArticle

Keha, Ahmet B. ; Duman, Tolga M. / Minimum distance computation of LDPC codes using a branch and cut algorithm. In: IEEE Transactions on Communications. 2010 ; Vol. 58, No. 4. pp. 1072-1079.
@article{441603b33fb5405aa31b769d422864dc,
title = "Minimum distance computation of LDPC codes using a branch and cut algorithm",
abstract = "We give a branch-and-cut algorithm for finding the minimum distance of a binary linear block code. We give two integer programming (IP) models and study the convex hull of the single constraint relaxation of these IP models. We use the new inequalities as cuts in a branch-and-cut scheme. Finally, we report computational results based on turbo and low density parity check (LDPC) codes that demonstrate the effectiveness of our cuts. We demonstrate that our IP formulation and specific cuts are efficient tools for determining the minimum distance of moderate size linear block codes, specifically, they are very efficient for LDPC codes, and provide us with an additional tool for solving this important problem.",
keywords = "Branchand-cut, Linear block codes, Low density parity checkcodes, Mixed-integer programming modelling, Turbo codes",
author = "Keha, {Ahmet B.} and Duman, {Tolga M.}",
year = "2010",
month = "4",
doi = "10.1109/TCOMM.2010.04.090164",
language = "English (US)",
volume = "58",
pages = "1072--1079",
journal = "IEEE Transactions on Communications",
issn = "0090-6778",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

TY - JOUR

T1 - Minimum distance computation of LDPC codes using a branch and cut algorithm

AU - Keha, Ahmet B.

AU - Duman, Tolga M.

PY - 2010/4

Y1 - 2010/4

N2 - We give a branch-and-cut algorithm for finding the minimum distance of a binary linear block code. We give two integer programming (IP) models and study the convex hull of the single constraint relaxation of these IP models. We use the new inequalities as cuts in a branch-and-cut scheme. Finally, we report computational results based on turbo and low density parity check (LDPC) codes that demonstrate the effectiveness of our cuts. We demonstrate that our IP formulation and specific cuts are efficient tools for determining the minimum distance of moderate size linear block codes, specifically, they are very efficient for LDPC codes, and provide us with an additional tool for solving this important problem.

AB - We give a branch-and-cut algorithm for finding the minimum distance of a binary linear block code. We give two integer programming (IP) models and study the convex hull of the single constraint relaxation of these IP models. We use the new inequalities as cuts in a branch-and-cut scheme. Finally, we report computational results based on turbo and low density parity check (LDPC) codes that demonstrate the effectiveness of our cuts. We demonstrate that our IP formulation and specific cuts are efficient tools for determining the minimum distance of moderate size linear block codes, specifically, they are very efficient for LDPC codes, and provide us with an additional tool for solving this important problem.

KW - Branchand-cut

KW - Linear block codes

KW - Low density parity checkcodes

KW - Mixed-integer programming modelling

KW - Turbo codes

UR - http://www.scopus.com/inward/record.url?scp=77950598865&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950598865&partnerID=8YFLogxK

U2 - 10.1109/TCOMM.2010.04.090164

DO - 10.1109/TCOMM.2010.04.090164

M3 - Article

AN - SCOPUS:77950598865

VL - 58

SP - 1072

EP - 1079

JO - IEEE Transactions on Communications

JF - IEEE Transactions on Communications

SN - 0090-6778

IS - 4

M1 - 5439310

ER -